Written Answers. - Social Welfare Survey.

Joe Walsh


87 Mr. J. Walsh asked the Taoiseach when the recent Central Statistics Office social welfare fraud survey was commissioned; the study objectives involved; and the duration and nature of assessment of its findings before it was made public. [17029/96]

The Government established a Strategy Group on Employment and Unemployment following publication of the Labour Force Survey — LFS — results last October. The group was asked, inter alia, to examine the reasons for the growing divergence between the live register and the LFS estimates of unemployment. During discussions of the options for effective analysis of this divergence, the group supported a proposal by the Central Statistics Office — whose input had been requested by the group — to conduct a statistical exercise as a quality check of the LFS estimate of unemployment. This special study was to be conducted in conjunction with the April 1996 LFS and it aimed to assess more directly the reasons for the difference between the live register and LFS estimates of unemployment. This was not a “fraud survey”. The CSO presented an interim report on the study to the Strategy Group on Employment and Unemployment on Wednesday, 11 September and published the main results of the study on Wednesday, 18 September.

The final figures published by the CSO on 18 September differed from those presented in the interim report. This was because: the interim figures were unweighted sample results whereas the final figures were weighted to reflect the overall distribution of the live register between long-term and short-term claimants; some categories who are not part of the live register count — such as systematic short-time workers — were included in the interim analysis but were subsequently excluded from the final figures; and the final figure was also adjusted for an estimated small level of outflows from the live register between 19 April 1996 and the reference period of the LFS. These refinements gave a more accurate assessment of the divergence between the LFS estimate of unemployment and the live register count.